Conservation prioritization of dry forest community types, and species in the central dry of Myanmar

Background/Question/Methods

Tropical dry forests are more threatened, less protected and susceptible to deforestation compared to tropical rain forests. Although dry forests in Myanmar are now in critical condition due to human disturbances, which vegetation community and species should be given priority for conservation have yet known due to limited resources for research and conservation activities. This study tries to develop conservation prioritization procedure by the feasible effort of field survey and available maps for conserving dry forest community types and species. Presence-absence survey for all vascular plants was done in 1139 sample plots of 15m x 15m in 100km x 100km study area. Community types were classified by two way indicator species analysis (TWINSPAN) with Chi- squared test. Community level modeling approach was applied with generalized linear model (GLM) for plant community types based on environmental predictors derived from DEM (GTOPO 30) and land use map produced by drawing in Google Earth. Decision procedure based on scoring human impact, extreme plant functional type, extreme environment and area remained was used to determine prioritized vegetation types for conservation. To identify prioritized species, log odds ratio method based on the importance value, appear at the target community, and rarity was applied.

Results/Conclusions

334 species were recorded, 14 community types were classified and their indicator species were identified. Predicated occurrence probability maps for 14 community types can be produced. Types 1 to 6 was found in remnant forests which mostly remain in high elevation and slope while others were found in agricultural land and residential area. 6 community types which cover 5.03% of the total study and 79 plant species as prioritized group A (38 species) and group B (41 species) can be determined for conservation prioritization. The predictive maps for vegetation communities based on community level modeling gave key information for assessing their distribution status, extent of occurrence and baseline data for cost effective conservation prioritization for regional biodiversity conservation planning. Land use and elevation are the most important factors explaining the community types. The 6 prioritized community types to be conserved proposed by decision analysis absolutely corresponds to three Myanmar experts judgment. Prioritized species detected by log odds ratio method also mostly correspond to expert judgment. The research findings indicate that the proposed method will be applicable not only to conservation prioritization of dry forest in Myanmar but also to other countries without enough data on regional biological diversity.